International peer-reviewed academic journals call for papers, http://www.iiste.org/Journals

1. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
131
Factors Affecting an Age at First Marriage among Female
Adolescents in Bangladesh
Abdul Hamid Chowdhury, Mohammad Nazmul Hoq, Mohammad Emdad Hossain ,
Md. Musa Khan
Department of Business Administration, International Islamic University Chittagong, Bangladesh
Abstract
The aim of this paper is to study the factors associated with age at first marriage among female adolescents
utilizing the nationally representative survey data of Bangladesh Demographic and Health Survey (BDHS),
2007. Logistic regression analyses have been used to study the factors affecting on age at first marriage in
Bangladesh. Respondent’s educations, husband’s education, region and working status of respondents are found
to have significant effect on age at first marriage. Findings need to be scientifically utilized in developing
suitable programs addressing the case of early marriage, particularly in Barisal, Khulna and Rajshahi.
Keywords: Adolescence, Age at first marriage, Contingency table, Odds ratio.
1. Introduction
Bangladesh has one of the world’s highest rates of early marriage. According to (UNICEF's flagship report,
2011) figures, 66 percent of Bangladeshi girls are married before the age of 18 and approximately a third of
women were married by the age of 15 although the legal age at first marriage for females in Bangladesh is 18
years. It is generally believed that rural tradition fosters early marriage, while urbanization and other forces of
modernization lead to marriage postponement. In recent decades, there has been a general rise in the average age
at first marriage in Asia, though the initial age levels and rate of change have varied considerably from country
to country. Early marriage which is also referred to as child marriage is common all over the globe and has
inflicted dangerous and devastating effects on young children (especially females) who are completed to tie the
knot in most cases. While the age at marriage is generally on the rise, early marriage of children and adolescents
below the age of 18 is still widely practiced (UNICEF Innocenti Research Centre, Innocenti Digestno. 7, 2001).
Early marriage of female children is rampant in Bangladesh, especially in slum areas, where there is a noticeable
lack of back infrastructure, services and basic shelter. Those areas are characterized by substandard housing and
squalor and lacking in tenure security. For female, the problem of early marriage is acute not only in slum areas
but throughout the country as well. State’s legal provision on minimum age at first marriage in Bangladesh
generally is 18 years for female to access in marriage relationship. The average age at first marriage for female
in Bangladesh is 18.7 years (Bangladesh Bureau of Statistics, BBS, 2009).
Early marriage is more frequent for female than their male counterparts. In rural Bangladesh there are many
social pressures to marry of pubescent girls (Aziz & Maloney, 1985). If the marriage of a pubescent girl is
delayed, her parents and sometimes the girl herself are made to feel guilty. Similar scenario is more common in
slum areas. Poverty is one of the major factors under pining early marriages. Where poverty is acute, a young
girl may be regarded as an economic burden and parents looking for early marriage of their daughters as an
alternative way of reducing the burden. Bangladesh is a developing country where industrialization starts to
begin in the recent year and industrialization leads to the urbanization. The growth of cities has been always
accompanied with the growth of slums, where a huge number of people live. Deteriorating socio-economic
conditions of slum dwellers is a major cause of unstable and fragile life of the women living there. Early
marriage also has implications for the well-being of families, and for society as a whole. It extends a women’s
potential childbearing capacity, which itself represents a risk to mother (Arjun, Ayad, & Kumar, 1991). Late
marriage always cut the reproductive span of a woman and marriage at an earlier age gives more time to produce
birth throughout her reproductive life span. Under these circumstances, early marriage may bring some physical
and mental complexities to the mothers. Progress made by Bangladesh regarding this issue, especially those
factors that are related to early marriage and early conception has extremely. However, no systematic effort to
slump down rate has been strongly undertaken by civil society organizations (CSOs) and public or private
organizations as well. Thus, this study is making a humble effort to make an assessment regarding early
marriage, so that, it can have a clear image of the current situation of the regarding matters in adolescent women
marriage in Bangladesh.

2. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
132
2. Objectives of the study
In this study, an attempt has been made to examine the predictors of age at first marriage in Bangladesh. There
are many factors that influencing on age at first marriage. Thus, a greater attention has to be paid to find out the
factors that are influencing on early marriage and ultimately the growth of population. However, the specific
objectives of this research are:
i. To find the percentage of age at first marriage of adolescent women by various socio-economic variables;
ii. To determine whether age at first marriage vary among adolescent women of Bangladesh by selected socio-
economic and demographic characteristics and
iii. To examine the effects of available socio-demographic factors on age at first marriage.
3. Data and Methodology
This study uses data extracted from the report of Bangladesh Demographic and Health Survey (National
Institute of Population Research and Training (NIPORT), 2009). The survey was conducted between 24 March
and 11 August 2007, on behalf of the Government of Bangladesh by National Institute for Population Research
and Training (NIPORT). The project was funded by the United States Agency for International Development
(USAID)/Dhaka. The sampling frame for the survey considered all households in Bangladesh from which a
nationally representative sample of 10,819 households was selected; 10,461 were occupied. Of the households
occupied, (10400) 99.40 percent were successfully interviewed. In these households, a total of 11,178 ever-
married females aged less than 50 years were identified as suitable for individual interview. Of them, 10996
females (or 98.4%) were successfully interviewed.
Among the 10996 ever-married females, the numbers of urban and rural respondents were 4151 (37.75%) and
6845 (62.25%), respectively. The sample had been taken 5 years prior to BDHS-2007 survey (National Institute
of Population Research and Training (NIPORT), 2009). Out of 10996 ever-married females, 1348 (12.26%) were
found at age under 20, known as adolescents. The associations between age at first marriage and selected
explanatory variables have been tested by applying cross-tabulation analysis. The cross-tabulation analysis was
important at first step for studying the relationship of age at first marriage with several characteristics. However,
such analysis fails to address age at first marriage predictors completely because of ignoring other covariates.
Hence, Logistic regression analysis has also been adopted in order to estimate independent effects of each
variable while controlled for others. This analysis has considered all the covariates that have found significant in
cross-tabulation analysis.
3.1 Linear logistic regression model
For a single variable, the logistic regression model is of the form,
Prob (event) =
x
e
x
e
101
10
ββ
ββ
+
+
+
Or, equivalently, Prob (event) =
)10(-
e1
1
xββ +
+
Where β0 and β1 are the regression coefficients estimated from the data and X is the independent variable.
For more than one variable, the model is

3. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
133
Prob (event) =
PxPβ........1x1β0β
e1
PxPβ........1x1β0β
e
+++
+
+++
=
ze
ze
+1
=
ze−+1
1
Where, z = β0 +β1 X1 +β2 X2 +………+βP XP
In this model, the parameters are estimated by the maximum likelihood method. Here the probability lies
between 0 and 1 while the relationship between the probability and the independent variables is non-linear. The
regression coefficients imply that how much change of dependent variable by a one-unit change of independent
variables. A statistic that is used to take at the partial correlation between the dependent variable and each of the
independent variables is the R statistic. The independent variables are fitted to the logistic model at a time by
stepwise selection procedure. The process continues to add new variable to the regression equation at each step
until the regression is satisfactory. The list of dependent variable and independent variables is shown by the
Table 1.
4. Analysis and Discussion
Percentage and Cross tabulation of adolescent respondents’ age at first marriage by different characteristics are
presented by Table 2. The dependent variable, age at first marriage is categorized into two classes i.e. one is,
before and at median age at first marriage (15 years) and another one is above 15 years. We use χ2 test for the
independency of attributes. From the Table-2, it is seen that all of the independent variable except religion come
out significant at bi-variate level. Table 2 reveals that respondent’s education has a strong significant relationship
with age at first marriage. Adolescent respondents who have no formal education married early. The data show
that higher educated respondents have tendency to marry later than that of primary educated adolescent
respondents. 43.8 percent higher educated adolescent respondents married after median age following by 25.8
and 25.2 percent of primary and illiterate respondents. Early marriage has highest proportion in Rajshahi division
among divisions, i.e. before or at median age 72.0 percent adolescent respondents married and the next higher
proportion is in khulna division (71.3 percent). Barisal (66.3 percent), Dhaka (61.6 percent), Chittagong (55.1
percent) and Sylhet (53.2 percent) are the successive descending order of proportions of marriage before or at
median age. This regional differentiation is observed because of industrialization, urbanization and education. It
is noticed from the table that the respondents who live in rural area have tendency to marry early than urban
respondents. 66.1 percent rural respondents married before or at median age with compare to 58.3 urban
respondents. Chi-square test at 0.1 percent level of significance shows that there is a strongly positive association
between place of residence and age at first marriage.
From the Table 2, we can expatiate on the occupation of husbands, which has significant effect on female
adolescents’ age at first marriage. Husband who has better occupation usually marry a woman who is conscious
about her life. The data reveals that 45.9 percent respondents whose husbands are in service married after median
age followed by business (45.7 percent), others (32.4 percent) and agriculture (30.2 percent). Husband’s
education is not as strong as women’s education but has significant association on respondent’s age at first
marriage. The respondents whose husbands are illiterate, 75.9 percent married on or before median age first
marriage than the respondents whose husbands have primary (70.4 percent) and higher (50.9 percent) level of
education. Because educated husbands are likely to marry educated females. Respondents’ currently working
status has also significant effect on age at first marriage. Table 2 shows that the respondents who are not
currently working married early than respondents who are working. 73.9 percent adolescent women who are not
currently working married before or at median age while 61.7 percent adolescent women who are currently
working married at the same interval. Access to mass media has highly significant effect on age at first marriage.
The reason is that the respondents who are come from rich family marry later age as compare to other status
family. The rich respondent get all the privilege from society like as education, access to mass media and other
facilities which directly affect age at first marriage. In this study, 54.2 percent rich adolescent women married
before or at median age while the percentages of other status are middle (67.5 percent) and poor (38.6 percent) in
the same interval.

4. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
134
Logistic regression model is fitted (Table 3) to the data to examine the effect of those independent variables
which are found significant in χ2 test. Table-3 gives the estimates of the logistic regression coefficients (β)
corresponding to the independent variables and relative odds for each categorical variable. Respondent’s
education is found to be statistically significant on age at first marriage. The odds ratio for the respondents who
have secondary and higher education is 1.899. This implies that the likelihood of getting married after 15 years
of adolescent women with secondary and higher education is 1.899 times higher than that of getting married
below or at 15 years of illiterate adolescent women. The analysis further shows that region has a strong
significant effect on age at first marriage. The odds ratio for the Chittagong and Sylhet are 1.686 and 2.438
respectively. It implies that the respondents who lived in Chittagong division are 1.686 times more likely to
marry at later ages as compared to the respondents who lived in Barisal division. Moreover, the respondents who
lived in Sylhet division are 2.438 times more likely to marry after age 15 years as compared to the respondents
who lived in Barisal division. Husband education is found to have a highly significant influence on age at first
marriage of respondents. The odds ratio for the husbands who have secondary and higher education indicates
that they are 2.116 times more likely to marry after 15 years than the respondents whose husbands have no
formal education. It is also observed from the table that occupation of respondent has statistically significant
effect on age at first marriage. Regression coefficient indicates that there is a positive relationship between
working women and age at first marriage. The odds ratio for working women indicates that they are 1.453 times
more likely to marry at later ages with compare to those who are not in working.
5. Conclusion
This study investigates the predictors of age at first marriage in Bangladesh among female adolescents. It has
used the national representative data from the Bangladesh Demographic and Health Survey (National Institute
of Population Research and Training (NIPORT), 2009). Both cross-tabulation and Logistic regression analysis
techniques have been applied to identify the important predictors of age at first marriage. Since age at first
marriage is one of the important proximate determinants of population size, the present study also analyzed the
age at first marriage by teenage females. Patterns of marriage show that about 63.7 percent respondents were
married up to age 15 years. The median age at first marriage is found to be 15 years, which is 3 years less than
legal age at first marriage of females in Bangladesh. The result suggests that early marriage among females is a
multidimensional phenomenon. Logistic regression analysis exhibits that explanatory variables such as
respondent’s education, husband’s education, region and respondents working status are important in explaining
differentials of age at marriage of the Bangladeshi adolescents. Of all the variables, respondent’s education and
husband’s education makes by far the strongest contribution to the variability in age at first marriage of the
adolescents. The findings of the present study have clear policy implications. To increase age at first marriage
and to check the growth rate of population, highest importance should be attached to education of the
respondents. If literacy rate can be increased it would develop a sense of national awareness and wide outlook
among them.
Hence, all-out efforts should be taken to weed out female’s illiteracy. Initiatives must also be taken to ensure at
least secondary education level among girls. In this context, possibility of free education for females up to
secondary level can be justified, which will accelerate the females towards higher ages at first marriage.
Husband’s level of education has impact on age at first marriage. Hence, it can be suggested that male’s
education must be improved which will help to reduce childhood marriage. Women who are currently working
are more aware of their marriage. For this reason, the Government should come forward to create job
opportunities for women. Geographical region of residence is found to be a significant predictor of age at first
marriage. Therefore, economic disparity should be reduced and uniform distribution of national income must be
ensured across the country. If the aforementioned recommendations can be implemented properly, the age of
first marriage of adolescent women will be increased at expected level and accordingly the rate of infant and
maternal mortality can be reduced, which occur due to conception in early age.
References
Arjun, A., Ayad, M. & Kumar, S., 1991. The Role of Nuptiality in Fertility Decline: A comparative analysis.
Washington DC, Columbia, Maryland, s.n.
Aziz, K. & Maloney, C., 1985. Life Stages, Gender and Fertility in Bangladesh, Dhaka: International Centre for
Diarrhoeal Disease Research, Bangladesh.

5. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
135
Bangladesh Bureau of Statistics (BBS), 2011. Bangladesh Population Census 2011, National Volume, Dhaka,
Bangladesh: Statistics Division, Ministry of Planning.
Bangladesh Bureau of Statistics, BBS, 2009. Statistical Pocket Book of Bangladesh, Dhaka: Ministry of
Planning, Government of the People’s Republic of Bangladesh, Statistics Division.
Ikamari, L., 2005. The effect of education on the timing of marriage in Kenya. Demographic Research, 12(1),
pp. 1-28.
IRIN News, 31 July 2009 (IRIN), DHAKA, BANGLADESH: Too young to marry. Available from:
HYPERLINK http://www.irinnews.org/printreport.aspx?reportid=85516 (Accessed 25 November 2012)
Jensen, R. & Thornton, R., 2003. Early Female Marriage in the Developing World. Gender and Development,
July.pp. 9-19.
Jones, G., 1992. Divorce in Islamic South-East Asia. Working Papers in Demography, Canberra: Australian
National University.
National Institute of Population Research and Training (NIPORT), 2009. Bangladesh Demographic and Health
Survey-2007, Dhaka, Bangladesh: Mitra and Associates, Dhaka, Bangladesh, Macro International, Calverton,
Maryland USA.
UNICEF Innocenti Research Centre, Innocenti Digestno. 7, 2001. Early and forced marriage DIGEST, Italy:
UNICEF Innocenti Research Centre in Florence.
UNICEF's flagship report, 2011. State of the World's Children 2011 to adolescents, s.l.: UNICEF.
VandenHeuvel, A. & McDonald, P., 1994. Marriage and divorce. Beginning Population Studies, In Lucas D,
Meyer P (eds.), pp. 69-79.
Zabin, L. & Kiragu, K., 1998. The health consequences of adolescent sexual and fertility behavior in sub-
Saharan Africa. Studies in Family Planning, 29(2), pp. 210-32.
First Author: Dr. Abdul Hamid Chowdhury, Associate Professor, Department of Business Administration,
International Islamic University Chittagong, Bangladesh, ahamidc@gmail.com},
Second Author: Mohammad Nazmul Hoq, Lecturer, Department of Business Administration, International
Islamic University Chittagong, Bangladesh ronyfirst@yahoo.com
Third Author: Mohammad Emdad Hossain, Associate Professor, Department of Business Administration,
International Islamic University Chittagong, Bangladesh, mehapstat@gmail.com
Fourth Author: Md. Musa Khan, Assistant Professor, Department of Business Administration, International
Islamic University Chittagong, Bangladesh, musa_stat@yahoo.com
Field of Research: Demography

6. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
136
Table 1: The list of dependent variable and independent variables
Dependent variable Age at first marriage less than or equal to 15 years is “0”
Otherwise (above age 15 years ) “1”
Independent Variables Categories
Respondent Education 0 = Illiterate
1 = Primary literate
2 = Secondary and higher
Religion 0= Muslim
1= Non-Muslim
Region 0= Barisal
1= Chittagong
2= Dhaka
3=Khulna
4= Rajshahi
5= Sylhet
Husband’s occupation 0= Agriculture
1= Business
2= Service
3= Others
Type of place of residence 0= Rural
1= Urban
Husband’s education 0= Illiterate
1= Primary literate
2= Secondary and higher
Access to mass media 0= no access
1= have access

7. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
137
Socio-economic status 0= Poor
1= Middle
2= Rich
Working status 0 = Not working
1 = Working
Table 2: Percentage and Cross tabulation of adolescent respondents’ age at first marriage by different
characteristics
Background Characteristics
Age at first marriage
at median age
Chi-square Significance Level
≤15 years >15 years
Respondent’s education
Illiterate
Primary literate
Secondary and higher
74.8
74.2
56.8
25.2
25.8
43.2
43.069 0.000
Religion
Muslim
Non-Muslim
64.2
57.7
35.8
42.3
1.623 0.123
Region
Barisal
Chittagong
Dhaka
Khulna
Rajshahi
Sylhet
66.3
55.1
61.6
71.3
72.0
53.2
33.7
44.9
38.4
28.7
28.0
46.8
29.904 0.000
Type of place of residence
Rural 66.1 33.9 7.519 0.006

8. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
138
Background Characteristics
Age at first marriage
at median age
Chi-square Significance Level
≤15 years >15 years
Urban 58.3 41.7
Husband’s occupation
Agriculture
Business
Service
Others
69.8
54.3
54.1
67.6
30.2
45.7
45.9
32.4
25.892 0.000
Husband’s education
Illiterate
Primary literate
Secondary and higher
75.9
70.4
50.9
24.1
29.6
49.1
70.721 0.000
Access to mass media
No access
Have access
72.3
60.2
27.7
39.8
17.491 0.000
Socio-economic status
Poor
Middle
Rich
72.2
67.5
54.2
27.8
32.5
45.8
38.875 0.000
Respondent currently working
No
Yes
73.9
61.7
26.1
38.3
11.844 0.001
Total 63.7(859) 36.3(489)

9. Research on Humanities and Social Sciences www.iiste.org
ISSN 2222-1719 (Paper) ISSN 2222-2863 (Online)
Vol.3, No.9, 2013
139
Table 3: Logistic regression estimates of age at first marriage by different characteristics
Factors Coefficient (ββββ) Odds ratio[exp(ββββ)]
Respondent’s education
Illiterate(RC)
Primary literate
Secondary and higher
-
0.109
0.641***
-
1.115
1.899
Region
Barisal(RC)
Chittagong
Dhaka
Khulna
Rajshahi
Sylhet
-
0523**
0.256
-0.233
-0.269
0.891***
-
1.686
1.291
0.792
0.764
2.438
Type of place of residence
Rural(RC)
Urban
-
0.112
-
1.119
Husband’s occupation
Agriculture(RC)
Business
Service
Others
-
0.094
0.260
-0.105
-
1.099
1.297
0.900
Husband’s education
Illiterate(RC)
Primary literate
Secondary and higher
-
0.113
0.750***
-
1.120
2.116
Access to mass media
No access(RC)
Have access
-
0.221
-
1.247
Socio-economic status
Poor(RC)
Middle
Rich
-
-0.116
0.077
-
0.890
1.080
Respondent currently working
No(RC)
Yes
-
0.374**
-
1.453
Constant -2.068*** 0.126

10. This academic article was published by The International Institute for Science,
Technology and Education (IISTE). The IISTE is a pioneer in the Open Access
Publishing service based in the U.S. and Europe. The aim of the institute is
Accelerating Global Knowledge Sharing.
More information about the publisher can be found in the IISTE’s homepage:
http://www.iiste.org
CALL FOR PAPERS
The IISTE is currently hosting more than 30 peer-reviewed academic journals and
collaborating with academic institutions around the world. There’s no deadline for
submission. Prospective authors of IISTE journals can find the submission
instruction on the following page: http://www.iiste.org/Journals/
The IISTE editorial team promises to the review and publish all the qualified
submissions in a fast manner. All the journals articles are available online to the
readers all over the world without financial, legal, or technical barriers other than
those inseparable from gaining access to the internet itself. Printed version of the
journals is also available upon request of readers and authors.
IISTE Knowledge Sharing Partners
EBSCO, Index Copernicus, Ulrich's Periodicals Directory, JournalTOCS, PKP Open
Archives Harvester, Bielefeld Academic Search Engine, Elektronische
Zeitschriftenbibliothek EZB, Open J-Gate, OCLC WorldCat, Universe Digtial
Library , NewJour, Google Scholar